Issue: August 2014
August 01, 2014
18 min read

Testosterone treatment patterns, risks spark passionate debate among experts

Issue: August 2014
You've successfully added to your alerts. You will receive an email when new content is published.

Click Here to Manage Email Alerts

We were unable to process your request. Please try again later. If you continue to have this issue please contact

With recent research suggesting cardiovascular risks with testosterone treatment, the debate continues among experts in the field as to the validity of the claims and the true risk–benefit analysis of treatment with the sex hormone.

As Abraham Morgentaler, MD, FACS, director and founder of Men’s Health Boston and associate clinical professor of urology at Harvard Medical School, told Endocrine Today, discussion of sex hormones, especially testosterone, often gets heated.

“People have a lot of passion about the sex hormones. Everyone has an opinion about testosterone and estrogen,” he said. “Testosterone therapy is not a panacea, but for men who are symptomatic and testosterone deficient, it’s a successful treatment.”

Morgentaler detailed many reasons for his treatment methods and why current research is not enough to change decades’ worth of practice, but other experts — who are in agreement that the research is not up to par — said there is still a lack of evidence of true safety and benefit.

Ranjith Ramasamy, MD, a urologist at Baylor College of Medicine, said specific treatment goals need to be discussed with each patient.

Source: Ranjith Ramasamy, MD.

 “We need to be clear which patients we’re talking about. Everyone agrees that we should treat men with testosterone if they have a clear, identifiable cause for testosterone deficiency. The issue is with treating older men with non-specific complaints who have testosterone levels at the lower end of the reference range.

“We don’t know if testosterone is safe nor do we know if it is effective in these men, which is as important,” Joel S. Finkelstein, MD, director of Massachusetts General Hospital Bone Density Center, said in an interview. “The absence of evidence of toxicity is not the same as the evidence of absence of toxicity.”

Whom to treat

Besides the debate of risks vs. benefits, Mathis Grossmann, MD, PhD, FRACP, head of men’s health at Austin Health in Melbourne, Australia, discussed the differences in practitioners’ treatment parameters during the Endocrine Society annual meeting.

He showed results of a study in which physicians were surveyed and reported varied methodologies in diagnosing low testosterone. For example, 25% of respondents would diagnose androgen deficiency based on a single low testosterone level, only 44% would request a fasting sample, only 50% were aware of their local testosterone assay methodology and only 50% requested a repeat morning test of testosterone level.

Experts referenced the Endocrine Society guidelines as the standard for determining who is “low” and needs treatment. The guidelines suggest treating symptomatic men after measuring morning total testosterone with a repeat measurement for confirmation. If those measurements are near the low end of normal, the Society suggests measurement of free or bioavailable testosterone, using validated assays.

In an interview with Endocrine Today, Paresh Dandona, MD, PhD, chief of endocrinology at the University of Buffalo School of Medicine and Biosciences, said free testosterone should be utilized more often than it is.

Paresh Dandona, MD, PhD

Paresh Dandona

“We used to measure total testosterone and that was the only measurement available,” Dandona said. “It’s free testosterone that needs to be measured and used for diagnosis.”

He said the catch with measuring free testosterone is that not all labs are equipped to handle it accurately.

Morgentaler agreed: “I also measure free testosterone and find that for many of my patients who may have ‘normal’ testosterone levels, their free testosterone levels are low. ... Those men are candidates for treatment, and there’s some evidence that we’ve published that those men do just as well as if their total testosterone is low.”


Morgentaler sets low levels at a total testosterone of 350 ng/dL or less or a calculated free testosterone of 100 pg/mL or less.

Morgentaler began using testosterone to treat sexual dysfunction in the late 1980s, he said, when treatment was reserved for men at castrate levels. He attributed this restriction to a fear of prostate cancer and a lack of knowledge of the benefits of testosterone therapy.

“What became clear to me, quickly, was that testosterone deficiency ... was not at all rare and that men could tell when their levels were up or when they were low,” Morgentaler said.

Grossmann said symptoms are not always due solely to low testosterone.

“Within the symptoms, there are some symptoms that are more suggestive than others,” Grossmann said in an interview with Endocrine Today.

He pointed to gynecomastia, loss of sexual hair or low libido as indicators of truly low testosterone, whereas fatigue and erectile dysfunction are less specific and often related to comorbidities.

“The true androgen deficiency is often underdiagnosed. Conditions such as Klinefelter’s — low testosterone from puberty — there’s good evidence that the majority of these men are not diagnosed in their lifetime,” Grossmann said. “The big irony is that true hypogonadism is underdiagnosed. So the men who really need testosterone aren’t getting it enough, whereas the men who don’t need it are overtreated.”

Method, duration of treatment

Many of the experts agreed that testosterone treatment is often based upon relief of symptoms and treatment method is mostly based on patient preference, but treatment goals should be incorporated.

Grossmann said all testosterone treatment has a placebo effect in the time immediately after delivery. Almost all patients will say they are improved, but levels may not reflect that improvement and the placebo effect may not last.

“Sometimes giving the treatment doesn’t necessarily mean the blood level increases. So you need to check and make sure you’re giving enough. If the level is high enough and the patient doesn’t feel better, you need to tell the patient, ‘Testosterone’s not the answer for you,’” Grossmann said.

Ranjith Ramasamy, MD, a urologist at Baylor College of Medicine, told Endocrine Today he discusses specific treatment goals with each patient.

“Importantly, I ask before I start therapy exactly what symptoms we are starting for: if it’s low libido, if it’s erectile dysfunction, if it’s lack of energy, if it’s fatigue,” he said. “These are not medications that can just be given and patients can be on it for life. We should figure out an endpoint and target stop date.”

Finkelstein said he gives patients 4 to 6 months of therapy to bring levels to the middle of the reference range.

“When a patient returns, if he can look me in the eye and say, ‘This is fabulous, I feel terrific. It’s much different than I felt before,’ then I will continue his treatment,” he said. “That almost never happens, however. So, then I stop the testosterone.”

In many studies looking at testosterone prescriptions, Grossmann said only 10% of patients continue with therapy after 1 year.

“That may tell you that perhaps it doesn’t work for them,” he said.

“As willing as we are to start testosterone therapy, we should have the same willingness to stop,” Ramasamy said. “Where we get into trouble is where we don’t follow those guidelines and indiscriminately start prescribing testosterone with no baseline, no target endpoint and, most importantly, lack of follow-up.”

Replacement vs. supplementation

Grossmann and Finkelstein made an important distinction between classical hypogonadism due to medical disease and symptomatic men whose levels are borderline low.

“If you treat [classical hypogonadism], you’re not giving a medication, you’re just replacing the body’s natural ingredients,” Grossmann said. “The other we don’t really know if it’s pharmacotherapy or mere replacement.”


He said as men age, they have characteristics similar to young hypogonadal men.

“They do tend to lose muscle mass. They tend to put on some fat. Their bone density deteriorates. And they have often sexual symptoms,” Grossmann said. “The question is if this mild drop in testosterone that occurs with aging is causative to the symptoms or are they both parallel and don’t causatively relate.”

It was originally thought that even if it is due to aging, it is worth supplementing that decline with pharmacotherapy, Grossmann said.

“But now we know that decline is probably not very pronounced if you have healthy aging,” he said, pointing to obesity, diabetes and other comorbidities as possible factors. Additionally, low testosterone is a “very good barometer of health” and the question remains as to whether increasing the hormone can improve health.

“The principle is to improve health rather than treat [low testosterone], but if health doesn’t improve and they have a poor quality of life, then we should try it,” Grossmann said.

Finkelstein said his concern comes from “men who have non-specific complaints such as fatigue and heard about ‘Low T’ on the radio and are looking for an easy way to treat a difficult problem,” but Morgentaler said sometimes those symptoms are worth treating.

“For the majority of men who have symptoms suggestive of testosterone deficiency and levels that are consistent with it, my experience is that testosterone therapy is highly effective,” Morgentaler said.

First-line therapy or last resort

When presented with a patient with diabetes or obesity, many experts suggest first treating the comorbidity and monitoring to see if there is an improvement in testosterone levels.

“Type 2 diabetes and obesity are the two major factors to male hypogonadism,” Dandona said. “When we published in 2004 for the first time that one-third of type 2 diabetic patients have low testosterone, no one believed us.”

In 2008, they showed one-quarter of those with obesity had low testosterone.

Grossmann and others acknowledged that while obesity and glucose control should be addressed, there is potential for underlying hypogonadism.

“You can’t assume that [testosterone] will go up. You need to check it after they lose some weight,” Grossmann said.

Dandona agreed, saying they screen all of their patients with diabetes and obesity for testosterone levels.

“The majority of clinicians are not doing that, and therefore this is underdiagnosed in many patients who would benefit from this,” Dandona said. “Our recent work has shown that testosterone replacement actually increases insulin sensitivity.

“There is an improvement in the patient’s alertness, energy levels and buildup of muscle,” he said. “In addition, you get a reduction in adiposity.”

Morgentaler said testosterone can be an impetus for weight loss.

“Losing large amounts of weight can improve serum testosterone, but that’s also very difficult to achieve. What we’ve seen is that testosterone works in interesting ways on the brain, part of which is to improve motivation,” he said. “We always recommend exercise and diet to men who are obese, but I have a very low threshold for also treating them with testosterone if their levels are low. ... I’ve seen quite a few men who tell me they’ve never before experienced such success with weight loss as they have while on testosterone therapy.”

Grossmann said there is still concern in prescribing testosterone to those who have obesity and diabetes.

“That shouldn’t be an excuse for just prescribing treatment. We don’t know whether the treatment actually helps or causes harm,” Grossmann said.

Risks: real, relative or created

Grossmann’s statement stems from a lack of high-level data on CV risks and long-standing concerns with prostate side effects.

Both Finkelstein and Ramasamy said although testosterone therapy appears safe for men treated for prostate cancer, there are still questions regarding benign prostate hyperplasia (BPH) and prostate cancer under active surveillance.


“To my knowledge, there is no credible evidence to date that testosterone causes prostate cancer, however, the issue has not been well studied. I am more concerned about causing an exacerbation of benign prostate disease. For example, you might take an older man with a little bit of BPH who’s tolerating it well and cause enough enlargement in the prostate to make him symptomatic,” Finkelstein said.

Joel S. Finkelstein, MD

Joel S. Finkelstein

Ramasamy said, “The biggest question comes of giving testosterone to men diagnosed with prostate cancer and under active surveillance. ... The literature very obviously has taken two separate stances, and practitioners and patients are in a limbo as to what to do and how to proceed.”

Morgentaler said, “The fear that raising testosterone will lead to rapid growth of benign or malignant prostate tissue has been present since the 1940s. ... In 2011, my colleagues and I reported for the first time the effects of testosterone therapy in men on active surveillance for prostate cancer. It was a small group of only 13 men, but all of them had documented prostate cancer and all underwent follow-up prostate biopsies. After a mean of 2.5 years, none of the men demonstrated progression of their prostate cancer. This was the first time anyone had bothered to look at what happened to the prostates of these men if they underwent testosterone treatment. It turned out the answer was ‘not much.’

“I believe that the greatest impetus for the rise [in testosterone prescriptions] has been relaxation of the fear that testosterone increases the risk of prostate cancer. This has been shown in surveys to have been the number one concern for physicians with testosterone in the past,” he said.

Ramasamy said he monitors prostate-specific antigen levels in anyone older than 45 years or with a family history of prostate cancer.

Today, though, the overlying debate with testosterone risk surrounds CV events due to a study appearing in JAMA in November, and subsequent studies that seemed to support the concern. The Vigen et al study concluded that in men who underwent coronary angiography with low serum testosterone level, the use of testosterone therapy was linked with increased risk of death, heart attack or ischemic stroke

“Everyone was living happily, until two or three recent studies appeared and threw us off the track,” Dandona said. “These are badly conducted studies and do not allow us to make clear cut conclusions, and there are enough other studies that do not show adverse effects.”

He said much of the data was confined to the elderly in “carelessly” designed studies with “major weaknesses.”

“Testosterone therapy as a risk for CV disease had been off the table as observational study after observational study had shown that men with low testosterone were the ones at increased risk for mortality compared to normal testosterone, and as evidence accumulated, particularly over the last 10 years, the testosterone therapy in testosterone-deficient men improved several of the major risk factors for CV disease, including fat mass, obesity, insulin resistance and glycemic control,” Morgentaler said. “That single paper became embraced by many individuals as evidence of risk of testosterone, even though it became clear very quickly that paper itself was flawed.”

Morgentaler and other testosterone researchers created the Androgen Study Group, which sent a letter to the editor of JAMA urging retraction of the study by Vigen et al. It was signed by more than 160 testosterone experts from 32 countries, and 29 US and international medical societies in the fields of endocrinology, sexual medicine and men’s health.

The Androgen Study Group has also provided its own analysis of the study by Vigen et al to the FDA and the European Medicines Agency. The FDA will be addressing cardiovascular concerns in a September meeting.


Michael Ho, MD, PhD, an author on the Vigen et al study, said, “We stand by our results [and] ... we would like to emphasize that, as clearly stated in the manuscript, this was an observational study and did not conclude causality. The study conclusion was the following in the manuscript: ‘Future studies including randomized controlled trials are needed to properly characterize the potential risks of testosterone therapy in men with comorbidities.’ We support additional studies including [randomized controlled trials] in this area.”

In a letter submitted to the FDA, the Androgen Study Group did its own literature review, Morgentaler said.

“We’re unable to find a single study that shows that high levels of testosterone are associated with worse mortality or that testosterone therapy worsens known risk factors such as fat mass, obesity, waist circumference, glycemic control,” he said.

Finkelstein agrees that the recent studies are not high quality, particularly true of studies related to CV disease, but much of what was previously in the literature was also not powered to prove safety of testosterone therapy, he said.

“We don’t know what the risk-to-benefit ratio is for testosterone therapy. That’s the bottom line. The papers that have come out in the recent months are all terribly flawed,” Finkelstein said. “In the absence of knowledge regarding safety, the bar must be placed higher in terms evidence of efficacy. If you don’t know if it’s harmful, you have to be darn sure it’s helpful. ... In the absence of that information, a conservative approach is warranted.”

Ramasamy said he currently monitors estradiol due to the conversion of testosterone to estradiol and the recent link to estradiol and deep venous thromboembolism, as well as hemoglobin and hematocrit levels, to ward off concerns about arterial thrombosis and myocardial infarction.

“We don’t really know what the benefit is and especially if you’re an older man with heart disease, generally the practical thing would be to be cautious,” Grossmann said.

Morgentaler said this caution is hurting his patients.

“This particular article not only changed scientific concepts and our conversation about testosterone, but has negatively impacted public health care,” Morgentaler said. “Men in my practice who did well with testosterone for years stopped taking testosterone because they were concerned they would get a heart attack or stroke. Cardiologists have told my patients to stop taking this treatment. Men who are frankly testosterone deficient and are candidates for testosterone therapy declined treatment because of this perception that it is risky from a cardiovascular standpoint. And it is based on absolutely nothing.”


Finding uniformity

Experts agreed that the way to come to a true consensus about testosterone therapy would be through a time-consuming and expensive randomized clinical trial that is unlikely to happen.

There are ongoing studies that will be helpful in directing practice, the experts said. Several experts pointed to a new testosterone trial recommended by the Institute of Medicine and funded by the National Institute on Aging, which has finished recruitment and is almost through a year’s observation.

“That at least will give us an idea about benefits and perhaps which type of people we should treat,” Grossmann said.

Ramasamy said, “It was designed to see if testosterone will improve erectile function, but they did study cardiovascular risks as a secondary endpoint.”

He added, “We definitely need to educate not just the primary care doctors but also the public on the benefits and risks of testosterone therapy. ... Testosterone therapy needs to be an informed decision between the physician and the patient.”

Perhaps this education would help, Morgentaler said.

“There is an enormous amount of testosterone deficiency in this country that until recently has gone undiagnosed largely due to a lack of awareness of symptoms,” Morgentaler said.

Finkelstein said these studies may help in choosing the patients to treat, but they are likely still not powered to definitively show efficacy or harm.

“It’s amazing how much we don’t know. Testosterone has been around since the mid-1930s ... and we don’t know whether it’s effective. We don’t know whether it’s safe; we don’t know what the levels are,” Finkelstein said. “We know almost nothing. It’s shocking that a drug can be around for 80 years and we can have such limited knowledge about its clinical effects.” — by Katrina Altersitz

Paresh Dandona, MD, PhD, can be reached at the University at Buffalo School of Medicine and Biosciences, Dent Tower, 3980 Sheridan Drive, Amherst, NY 14226; email:
Joel S. Finkelstein, MD, can be reached at Endocrine Associates, 15 Parkman St., Boston, MA 02114-3117; email:
Mathis Grossmann, MD, PhD, FRACP, can be reached at Austin Hospital, 145 Studley Road, PO Box 5555, Heidelberg, Victoria, Australia 3084; email:
Michael Ho, MD, PhD, can be reached at Cardiology 111B, 1055 Clermont Street, Denver CO 80220.
Abraham Morgentaler, MD, FACS, can be reached at Men’s Health Boston, One Brookline Place, Suite 624, Brookline, MA 02445 email:
Ranjith Ramasamy, MD, can be reached at Baylor College of Medicine Medical Center, One Baylor Plaza, N 730, Houston, TX 77030; email:
Disclosure: Morgentaler has financial relationships with AbbVie, Antares, Auxilium, Clarus, Lilly and Lipocine. Dandona, Finkelstein, Grossmann and Ramasamy have no relevant financial disclosures.
Abdulmaged M. J Sex Med. 2014;11:624–629.
Bhasin S. J Clin Endocrinol Metab. 2010;95:2536-2559.
Grossmann M. J Clin Endocrinol Metab. 2011;96:2341–2353.
Dandona P. Curr Mol Med. 2008;8:816-828.
Dhindsa S. J Clin Endocrinol Metab. 2004;89:5462-5468.
Morgentaler M. J Urol. 2011;185:1256-1261.
Saad F. Obesity. 2013; doi:10.1002/oby.20407.
Vigen R. JAMA. 2013;310:1829-1836.


Are women who complain of sexual dysfunction candidates for testosterone therapy?


Women end up picking and choosing their own destiny

I have hundreds of patients who are women on testosterone, but women who have sexual dysfunction need to undergo a multidisciplinary evaluation.

If a woman with sexual dysfunction comes to our office, she spends roughly 4 hours at an initial visit. She will see a physical therapist, a sex therapist and then myself. She undergoes both a psychological and musculo-skeletal assessment and a full examination, including a vulvoscopy and vestibular examination, and a hormonal assessment.

Irwin Goldstein, MD

Irwin Goldstein


If patients, among their complaints, have a low sexual drive, poor sexual arousal, muted form of orgasm or inability to have orgasm, we will look at their testosterone.

We measure total testosterone, sex hormone binding globulin, dihydrotestosterone, estradiol, progesterone, the gonadotropins (LH and FSH), prolactin and thyroid. We will assess their endocrine status in robust ways both by physical symptoms and by blood test.

Somebody whose testosterone is in the lower tertile, whose sex hormone binding globulin is three to seven times the normal value of 25 nmol/L and whose calculated free testosterone is below a value that we have found to be normal in a woman without sexual dysfunction — around 0.6 to 0.8 ng/dL — we will be suspicious of a low testosterone.

If we think testosterone treatment is needed, we will review risks and benefits with the patients. We have a slideshow we show our patients. We will talk about data that shows safety and efficacy in pre- and post-menopausal women (Davis SR. J Sex Med. 2012;9(4):1134-48; Goldstat R. Menopause. 2003;10(5):390-8.) and data on testosterone patch safety and efficacy (Shifren JL. Menopause. 2006;13(5):770-9.).

We will explain that the FDA has not approved any form of testosterone for women and that use would be experimental, but we believe there to be sufficient data for women to make an informed decision.

If patients want to speak to other women on testosterone, we have many hundreds of patients who have done extremely well on testosterone and whose lives have been improved, whose life qualities have been improved. Many of these women are happy act as counselors to women who have questions.

At the end of the day, we offer three choices of delivery system: pellet, gel or intramuscular injection. We use the FDA-approved testosterones for men for these women, typically starting at 10% of the male dose and then adjust upwards or downwards based on their blood test values.

Testosterone is an amazing hormone. We have data that testosterone improves lean body mass, osteoporosis, C-reactive protein, waist circumference and muscle strength.

There are new data on testosterone and cognition and protection against dementia (Davis SR. Clin Endocrinol (Oxf). 2014; doi: 10.1111/cen.12459; Davis SR. Menopause. 2014;21(4):410-4.). As side effects we are looking at issues like acne, alopecia, and chin hairs vs. huge metabolic benefits.

Managing women with sexual health concerns requires multidisciplinary care including hormone treatment and among the hormones that we use is testosterone simply because epidemiologic studies show that at age 30 and going forward, their testosterone value falls.

Women end up picking and choosing their own destiny and we support their choices. Again, we’re not a group that’s purely biologic. In our facility, unlike most facilities, we have a sex therapist, physical therapist and myself, a physician. Patients are managed by all of us.

Irwin Goldstein, MD, is the president and director of The Institute for Sexual Medicine in San Diego.Disclosure: Goldstein has no relevant financial disclosures.


No testosterone level will ensure a good sex life

There’s really no need to check testosterone levels, as many well-designed studies have shown no relationship between testosterone levels and sexual dysfunction. In addition, all midlife women have “low” testosterone levels due to age alone.

There’s a tendency to pull out a prescription and that’s not the best way to approach this problem. It’s important to discuss and address common causes of sexual dysfunction before considering a trial of testosterone therapy. These may include fatigue, stress, limited ‘quality’ time with partner (eg, date nights), relationship conflict, partner performance and technique (eg, erectile dysfunction), vaginal dryness, bothersome hot flashes, depression, anxiety, and antidepressant therapy. Often addressing these issues can make a big difference in couples’ sex lives.


Jan L. Shifren, MD

Jan L. Shifren

I advise women or couples to see a sex therapist before considering any pharmacologic intervention because we as clinicians might not have the expertise to help women or couples communicate better about sex, to teach techniques to make sex more pleasurable, or to address past abuse or other concerns.

I’ve never had a woman or couple say sex therapy wasn’t helpful. I encourage physicians to identify good sex therapists in their communities. We wouldn’t think of treating diabetes without the help of a nutritionist, so we shouldn’t consider managing sexual concerns without the help of a good sex therapist.

After addressing all other causes of decreased libido, menopausal women with low desire associated with distress may consider a trial of off-label topical testosterone, though it’s important to inform women of limited efficacy and unknown long-term risks.

In our testosterone patch trials, total satisfying sexual events in a 4-week period increased by about one with placebo and about two with testosterone. In a recent (unpublished) large, randomized controlled trial of testosterone gel in postmenopausal women with sexual dysfunction, there was no significant improvement in libido or satisfying sexual events compared with placebo, despite increasing circulating testosterone to levels similar to those seen with the testosterone patch. This likely was because the placebo effect was very high, proving that much of libido resides above the neck.

Women must be informed that potential risks of testosterone therapy include hirsutism, acne, and adverse changes in lipids and liver function tests. In addition, there possibly may be an increased risk of estrogen-related adverse events, including breast cancer, venous thromboembolism, and coronary artery disease, as androgens are converted to estrogens. They also should know that this is not an FDA-approved therapy.

If a woman has no other cause of her sexual concerns and truly wishes a trial of testosterone, understanding limited efficacy, large placebo effect, and absence of long-term safety data, one option is compounded testosterone 1% cream or gel, 0.5 g applied topically nightly. Women must be informed that there are essentially no data on safety or efficacy for compounded testosterone. Total testosterone and SHBG levels must be checked on topical therapy, not as an efficacy measure (as no testosterone level will ensure a good sex life), but as a safety measure.

Placebos work very well, so instead of writing a prescription for testosterone, consider writing one for a “Date Night” instead.

Jan L. Shifren, MD, is director of the Massachusetts General Midlife Women’s Health Center in Boston. Disclosure: Shifren has no relevant financial disclosures.